qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Stefan Hajnoczi <stefanha@redhat.com>
To: qemu-devel@nongnu.org
Cc: Kevin Wolf <kwolf@redhat.com>,
	Paolo Bonzini <pbonzini@redhat.com>,
	Anthony Liguori <aliguori@us.ibm.com>,
	Ping Fan Liu <pingfank@linux.vnet.ibm.com>,
	Stefan Hajnoczi <stefanha@redhat.com>
Subject: [Qemu-devel] [RFC 07/13] block: add thread_aio_context TLS variable
Date: Fri, 14 Jun 2013 11:48:27 +0200	[thread overview]
Message-ID: <1371203313-26490-8-git-send-email-stefanha@redhat.com> (raw)
In-Reply-To: <1371203313-26490-1-git-send-email-stefanha@redhat.com>

BH and fd handler APIs need to know which AioContext (event loop) to run
inside.  Passing it around everywhere is not feasible since it would
require adding arguments to any call-chain that invokes BH and fd
handler APIs (hint: there are many and they change).

Instead make the AioContext pointer thread-local.  This way, any
function that needs to use an AioContext can find it.  In particular:
the iothread and vcpu threads share the main AioContext since they hold
the global mutex while running.  Dataplane threads will use their own
AioContexts.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
---
 async.c             | 2 ++
 cpus.c              | 2 ++
 include/block/aio.h | 4 ++++
 main-loop.c         | 1 +
 4 files changed, 9 insertions(+)

diff --git a/async.c b/async.c
index 90fe906..765cbc0 100644
--- a/async.c
+++ b/async.c
@@ -27,6 +27,8 @@
 #include "block/thread-pool.h"
 #include "qemu/main-loop.h"
 
+DEFINE_TLS(AioContext*, thread_aio_context);
+
 /***********************************************************/
 /* bottom halves (can be seen as timers which expire ASAP) */
 
diff --git a/cpus.c b/cpus.c
index c232265..529d612 100644
--- a/cpus.c
+++ b/cpus.c
@@ -741,6 +741,7 @@ static void *qemu_kvm_cpu_thread_fn(void *arg)
     qemu_thread_get_self(cpu->thread);
     cpu->thread_id = qemu_get_thread_id();
     cpu_single_env = env;
+    *alloc_thread_aio_context() = qemu_get_aio_context();
 
     r = kvm_init_vcpu(cpu);
     if (r < 0) {
@@ -815,6 +816,7 @@ static void tcg_exec_all(void);
 static void tcg_signal_cpu_creation(CPUState *cpu, void *data)
 {
     cpu->thread_id = qemu_get_thread_id();
+    *alloc_thread_aio_context() = qemu_get_aio_context();
     cpu->created = true;
 }
 
diff --git a/include/block/aio.h b/include/block/aio.h
index 1836793..6ee9369 100644
--- a/include/block/aio.h
+++ b/include/block/aio.h
@@ -17,6 +17,7 @@
 #include "qemu-common.h"
 #include "qemu/queue.h"
 #include "qemu/event_notifier.h"
+#include "qemu/tls.h"
 
 typedef struct BlockDriverAIOCB BlockDriverAIOCB;
 typedef void BlockDriverCompletionFunc(void *opaque, int ret);
@@ -74,6 +75,9 @@ typedef struct AioContext {
 /* Returns 1 if there are still outstanding AIO requests; 0 otherwise */
 typedef int (AioFlushEventNotifierHandler)(EventNotifier *e);
 
+/* Each thread can have a default AioContext */
+DECLARE_TLS(AioContext*, thread_aio_context);
+
 /**
  * aio_context_new: Allocate a new AioContext.
  *
diff --git a/main-loop.c b/main-loop.c
index cf36645..51eeb0f 100644
--- a/main-loop.c
+++ b/main-loop.c
@@ -142,6 +142,7 @@ int qemu_init_main_loop(void)
 
     gpollfds = g_array_new(FALSE, FALSE, sizeof(GPollFD));
     qemu_aio_context = aio_context_new();
+    *get_thread_aio_context() = qemu_aio_context;
     src = aio_get_g_source(qemu_aio_context);
     g_source_attach(src, NULL);
     g_source_unref(src);
-- 
1.8.1.4

  parent reply	other threads:[~2013-06-14  9:49 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-06-14  9:48 [Qemu-devel] [RFC 00/13] dataplane: use block layer Stefan Hajnoczi
2013-06-14  9:48 ` [Qemu-devel] [RFC 01/13] block: fix bdrv_flush() ordering in bdrv_close() Stefan Hajnoczi
2013-06-14  9:48 ` [Qemu-devel] [RFC 02/13] dataplane: sync virtio.c and vring.c virtqueue state Stefan Hajnoczi
2013-06-14  9:48 ` [Qemu-devel] [RFC 03/13] block: add BlockDevOps->drain_threads_cb() Stefan Hajnoczi
2013-06-14  9:48 ` [Qemu-devel] [RFC 04/13] virtio-blk: implement BlockDevOps->drain_threads_cb() Stefan Hajnoczi
2013-06-14 14:23   ` Paolo Bonzini
2013-06-14  9:48 ` [Qemu-devel] [RFC 05/13] exec: do not use qemu/tls.h Stefan Hajnoczi
2013-06-14  9:48 ` [Qemu-devel] [RFC 06/13] qemu-thread: add TLS wrappers Stefan Hajnoczi
2013-06-20  7:26   ` Fam Zheng
2013-06-20  8:50     ` Paolo Bonzini
2013-06-20 12:54       ` Stefan Hajnoczi
2013-06-14  9:48 ` Stefan Hajnoczi [this message]
2013-06-14  9:48 ` [Qemu-devel] [RFC 08/13] block: drop bdrv_get_aio_context() Stefan Hajnoczi
2013-06-14 14:12   ` Paolo Bonzini
2013-06-17 14:57     ` Stefan Hajnoczi
2013-06-17 15:03       ` Paolo Bonzini
2013-06-18  9:29         ` Stefan Hajnoczi
2013-06-14  9:48 ` [Qemu-devel] [RFC 09/13] main-loop: use thread-local AioContext Stefan Hajnoczi
2013-06-14  9:48 ` [Qemu-devel] [RFC 10/13] block: disable I/O throttling outside main loop Stefan Hajnoczi
2013-06-14  9:48 ` [Qemu-devel] [RFC 11/13] dataplane: use block layer for I/O Stefan Hajnoczi
2013-06-14  9:48 ` [Qemu-devel] [RFC 12/13] dataplane: drop ioq Linux AIO request queue Stefan Hajnoczi
2013-06-14  9:48 ` [Qemu-devel] [RFC 13/13] block: drop raw_get_aio_fd() Stefan Hajnoczi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1371203313-26490-8-git-send-email-stefanha@redhat.com \
    --to=stefanha@redhat.com \
    --cc=aliguori@us.ibm.com \
    --cc=kwolf@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=pingfank@linux.vnet.ibm.com \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).